Abstract:

Multi-modal hybrid index optimization is a kind of multi-objective optimization problems which is difficult to solve. This paper proposes an interactive cultural algorithm which has blended history knowledge, normative knowledge and domain knowledge according to double layer structures of cultural algorithm. This algorithm constructs sample database based on individual index equilibrium and chooses multi-modal solutions according to individual special crowding distance in decision space and objective space. These multi-modal solutions are recommend to user and as population clustering centers. According to population index equilibrium, knowledge guide adaptive crossover and mutation probability which can expand population diversity. This algorithm is applied to indoor layout optimization problem, and its outstanding performance is experimentally demonstrated.

Key words: evolutionary computation, cultural algorithm, interactive, hybrid index, optimization

摘要：

多模态混合指标优化是一类难以求解的多目标优化问题。针对该问题，借鉴文化算法的双层结构，构建了一种能融合历史知识、标准化知识和领域知识的交互式文化算法。该算法以指标均衡性构建信度空间样本库。知识提取函数根据样本库内个体在决策空间和目标空间的特殊拥挤距离选取多模态解。将选取的多模态解作为聚类中心推荐给用户评价。根据种群的指标均衡性，知识引导自适应交叉和变异概率，扩大种群多样性。采用指标均衡性引导形势知识更新。基于个体表现型相似性估计大规模种群隐式性能指标。提出新的多模态解评价测度。将算法应用于室内布局优化问题，与代表性方法比较，验证所提算法的有效性和可用性。

关键词: 进化算法, 文化算法, 交互, 混合指标, 优化